papyrus-scripts 2.1.0__tar.gz → 2.1.2__tar.gz

{papyrus_scripts-2.1.0/src/papyrus_scripts.egg-info → papyrus_scripts-2.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: papyrus_scripts
-Version: 2.1.0
+Version: 2.1.2
 Summary: A collection of scripts to handle the Papyrus bioactivity dataset
 Home-page: https://github.com/OlivierBeq/Papyrus-scripts
 Author: Olivier J. M. Béquignon - Brandon J. Bongers - Willem Jespers
@@ -17,7 +17,7 @@ Classifier: Programming Language :: Python :: 3.7
 Classifier: Programming Language :: Python :: 3.6
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy
+Requires-Dist: numpy>=2.0.0
 Requires-Dist: pandas
 Requires-Dist: rdkit
 Requires-Dist: requests
@@ -36,6 +36,7 @@ Requires-Dist: sphinx-autodoc-typehints; extra == "docs"
 Provides-Extra: testing
 Requires-Dist: pytest; extra == "testing"
 Requires-Dist: parameterized; extra == "testing"
+Dynamic: license-file
 
 # Papyrus-scripts
 
@@ -168,10 +169,10 @@ Different online servers host the Papyrus data based on release and ChEMBL versi
  
 | Papyrus version | ChEMBL version |                          Zenodo                           |                            4TU                            |
 |:---------------:|:--------------:|:---------------------------------------------------------:|:---------------------------------------------------------:|
-|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10944245) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
-|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/10943207) |                            :x:                            |
-|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7821775)  |                            :x:                            |
-|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787634) |                            :x:                            |
+|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10943992) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
+|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/7019873)  |                            :x:                            |
+|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7373213)  |                            :x:                            |
+|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787633) |                            :x:                            |
 
 Precomputed molecular and protein descriptors along with molecular structures (2D for default set and 3D for low quality set with stereochemistry) are not available for version 05.4 from 4TU but are from Google Drive.
 

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/README.md

@@ -129,10 +129,10 @@ Different online servers host the Papyrus data based on release and ChEMBL versi
  
 | Papyrus version | ChEMBL version |                          Zenodo                           |                            4TU                            |
 |:---------------:|:--------------:|:---------------------------------------------------------:|:---------------------------------------------------------:|
-|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10944245) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
-|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/10943207) |                            :x:                            |
-|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7821775)  |                            :x:                            |
-|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787634) |                            :x:                            |
+|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10943992) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
+|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/7019873)  |                            :x:                            |
+|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7373213)  |                            :x:                            |
+|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787633) |                            :x:                            |
 
 Precomputed molecular and protein descriptors along with molecular structures (2D for default set and 3D for low quality set with stereochemistry) are not available for version 05.4 from 4TU but are from Google Drive.
 

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/setup.cfg

@@ -32,7 +32,7 @@ packages = find:
 package_dir = 
 	= src
 install_requires = 
-	numpy
+	numpy>=2.0.0
 	pandas
 	rdkit
 	requests

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/src/papyrus_scripts/__init__.py

@@ -20,4 +20,4 @@ from .utils.IO import PapyrusVersion
 
 from .oop import PapyrusDataset
 
-__version__ = '2.1.0'
+__version__ = '2.1.2'

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/src/papyrus_scripts/oop.py

@@ -5,6 +5,7 @@
 from __future__ import annotations
 
 import os
+from abc import ABC
 from typing import Any, Dict, Iterator, List, Union, Optional
 
 import pystow
@@ -26,6 +27,15 @@ class PapyrusDataset:
     def __init__(self, version: str | IO.PapyrusVersion = 'latest', is3d: bool = False, plusplus: bool = True,
                  chunksize: Optional[int] = 1_000_000, source_path: Optional[str] = None,
                  download_progress: bool = False):
+        """Read, filter and aggregate data from a release of the Papyrus dataset.
+
+        :param version: version to be used. Either a `PapyrusVersion` or a `str` object to be passed to `PapyrusVersion` (default : 'latest')
+        :param is3d: should the lower-quality data with stereochemistry be read (default: False)
+        :param plusplus: read the Papyrus++ curated subset of even higher quality (defulat: True)
+        :param chunksize: number of lines per chunk. To read without chunks (not recommended) set to None (default: 1_000_000)
+        :param source_path: folder containing the bioactivity dataset (default: pystow's home folder)
+        :param download_progress: if the data not be on disk, should progress of the download be shown
+        """
         version = IO.PapyrusVersion(version=version)
         if not IO.is_local_version_available(version=version.version_old_fmt, root_folder=source_path):
             download.download_papyrus(outdir=source_path, version=version.version_old_fmt, nostereo=True, stereo=True,
@@ -48,13 +58,24 @@ class PapyrusDataset:
                        is3d: bool, version: str,
                        plusplus: bool = True,
                        source_path: Optional[str] = None,
-                       download_progress: bool = False
+                       download_progress: bool = False,
+                       chunksize: int = None
                        ) -> PapyrusDataset:
+        """Create a PapyrusDataset from a pandas DataFrame.
+
+        :param df: pandas DataFrame containing filtered Papyrus data samples (must preserve all default columns)
+        :param is3d: are the samples derived from the lower-quality data with stereochemistry
+        :param version: version of the dataset the samples were obtained from
+        :param plusplus: whether the samples were derived from Papyrus++
+        :param source_path: folder containing the original bioactivity dataset (default: pystow's home folder)
+        :param download_progress: if the data was not on disk, was progress of the download shown
+        :return: the `PapyrusDataset` object corresponding to the given samples
+        """
         dataset = PapyrusDataset.__new__(PapyrusDataset)
         dataset.papyrus_bioactivity_data = df
         dataset.papyrus_protein_data = reader.read_protein_set(source_path=source_path, version=version)
         dataset.papyrus_params = dict(is3d=is3d, version=version, plusplus=plusplus,
-                                      chunksize=len(df), source_path=source_path, num_rows=len(df),
+                                      chunksize=chunksize, source_path=source_path, num_rows=len(df),
                                       download_progress=download_progress)
         dataset._can_reset = False
         return dataset
@@ -64,6 +85,14 @@ class PapyrusDataset:
                    papyrus_protein_data: pd.DataFrame,
                    papyrus_params: Dict
                    ) -> PapyrusDataset:
+        """Create a PapyrusDataset from a bioactivities, proteins and initial parameters.
+
+        :param papyrus_bioactivity_data: bioactivity information obtained from a PapyrusDataset object
+        :param papyrus_protein_data: protein information obtained from a PapyrusDataset object
+        :param papyrus_params: parameters of the PapyrusDataset object `papyrus_bioactivity_data` and
+        `papyrus_protein_data` were obtained from
+        :return: the `PapyrusDataset` object corresponding to the given data
+        """
         dataset = PapyrusDataset.__new__(PapyrusDataset)
         dataset.papyrus_bioactivity_data = papyrus_bioactivity_data
         dataset.papyrus_protein_data = papyrus_protein_data
@@ -73,12 +102,14 @@ class PapyrusDataset:
 
     @property
     def _filter(self) -> PapyrusDataFilter:
+        """Create a PapyrusDataFilter object around the current dataset."""
         return PapyrusDataFilter(papyrus_bioactivity_data=self.papyrus_bioactivity_data,
                                  papyrus_protein_data=self.papyrus_protein_data,
                                  papyrus_params=self.papyrus_params)
 
     @property
     def _fpsubsim2(self) -> FPSubSim2Engine:
+        """Obtain the FPSubSim2Engine for substructure and similarity searches."""
         if self._fpsubsim2_ is None:
             self._fpsubsim2_ = FPSubSim2Engine(self.papyrus_params)
         self._fpsubsim2_._set_data(papyrus_bioactivity_data=self.papyrus_bioactivity_data,
@@ -86,56 +117,121 @@ class PapyrusDataset:
         return self._fpsubsim2_
 
     def keep_quality(self, min_quality: str) -> PapyrusDataset:
+        """Keep samples whose quality is at least the one supplied (e.g. 'medium' for both medium and high-quality)."""
         return self._filter.keep_quality(min_quality=min_quality)
 
     def keep_source(self, source: Union[List[str], str]) -> PapyrusDataset:
+        """Keep samples of specific data source(s) (e.g. 'chembl' or ['chembl', 'klaeger'])."""
         return self._filter.keep_source(source=source)
 
     def keep_activity_type(self, activity_types: Union[List[str], str]) -> PapyrusDataset:
+        """Keep samples of specific activity type(s) (e.g. 'ic50' or ['ki', 'ec50'])."""
         return self._filter.keep_activity_type(activity_types=activity_types)
 
     def keep_accession(self, accession: Union[List[str], str] = 'all') -> PapyrusDataset:
+        """Keep samples of specific accession(s) (e.g. 'P00533' or ['P11362', 'P35968'])."""
         return self._filter.keep_accession(accession=accession)
 
     def keep_protein_class(self, classes: Optional[Union[dict, List[dict]]],
                            generic_regex: bool = False) -> PapyrusDataset:
+        """Keep samples whose protein targets belong to a specific protein class(es).
+
+        :param classes: protein class(es) (e.g. {'l2': 'Kinase'} or [{'l2': 'Kinase'}, {'l1': 'Membrane receptor'}]).
+        :param generic_regex: should the generic pattern 'l?' be considered as a regex, allowing for partial matching.
+        """
         return self._filter.keep_protein_class(classes=classes, generic_regex=generic_regex)
 
     def keep_organism(self, organism: Optional[Union[str, List[str]]],
                       generic_regex: bool = False) -> PapyrusDataset:
+        """Keep samples whose protein targets belong to specific organisms.
+
+        :param organism: organism (e.g. 'Homo sapiens (Human)' or ['Bos taurus (Bovine)', 'Rattus norvegicus (Rat)'])
+        :param generic_regex: should partial matching be considered (e.g. 'human' or 'Pig')
+        """
         return self._filter.keep_organism(organism=organism, generic_regex=generic_regex)
 
     def contains(self, column: str, value: str, case: bool = True, regex: bool = False) -> PapyrusDataset:
+        """Keep samples with the specified field corresponding to the given value.
+
+        :param column: column to be checked for `value` in order for samples to be included
+        :param value: value the column must match
+        :param case: should the value matching be case-sensitive (default: True)
+        :param regex: should the given `value` be interpreted as a regular expression
+        """
         return self._filter.contains(column=column, value=value, case=case, regex=regex)
 
     def not_contains(self, column: str, value: str, case: bool = True, regex: bool = False) -> PapyrusDataset:
+        """Keep samples whose specified field not corresponding to the given value (opposite of the `contains` method).
+
+        :param column: column to be checked for `value` in order for samples to be included
+        :param value: value the column must match
+        :param case: should the value matching be case-sensitive (default: True)
+        :param regex: should the given `value` be interpreted as a regular expression
+        """
         return self._filter.not_contains(column=column, value=value, case=case, regex=regex)
 
     def isin(self, column: str, values: Union[Any, List[Any]]) -> PapyrusDataset:
+        """Keep samples whose value of the specified field is in the given options.
+
+        :param column: column to be checked for `values` in order for samples to be included
+        :param values: values the column must contain
+        """
         return self._filter.isin(column=column, values=values)
 
     def not_isin(self, column: str, values: Union[Any, List[Any]]) -> PapyrusDataset:
+        """Keep samples whose value of the specified field is not in the given options.
+
+        :param column: column to be checked for `values` in order for samples to be included
+        :param values: values the column must not contain
+        """
         return self._filter.not_isin(column=column, values=values)
 
     def keep_similar_molecules(self, smiles: Union[str, List[str]],
                                fingerprint: fingerprint.Fingerprint = fingerprint.MorganFingerprint(),
                                threshold: float = 0.7, cuda: bool = False) -> PapyrusDataset:
+        """Keep samples whose molecular structures are similar to any of the given SMILES.
+
+        :param smiles: SMILES the molecular structures must be similar to
+        :param fingerprint: type of fingerprint (subclass of `papyrus_scripts.fingerprint.Fingerprint`; default: `papyrus_scripts.fingerprint.MorganFingerprint`)
+        :param threshold: threshold of similarity to one of the given SMILES for a molecule to be considered a hit (default: 0.7)
+        :param cuda: should CUDA acceleration be used (default: False)
+        """
         return self._fpsubsim2.keep_similar_molecules(smiles=smiles, fingerprint=fingerprint, threshold=threshold,
                                                       cuda=cuda)
 
     def keep_dissimilar_molecules(self, smiles: Union[str, List[str]],
                                   fingerprint: fingerprint.Fingerprint = fingerprint.MorganFingerprint(),
                                   threshold: float = 0.7, cuda: bool = False) -> PapyrusDataset:
+        """Keep samples whose molecular structures are not similar to any of the given SMILES.
+
+        :param smiles: SMILES the molecular structures must not be similar to
+        :param fingerprint: type of fingerprint (subclass of `papyrus_scripts.fingerprint.Fingerprint`; default: `papyrus_scripts.fingerprint.MorganFingerprint`)
+        :param threshold: threshold of similarity to one of the given SMILES for a molecule to be considered a hit (default: 0.7)
+        :param cuda: should CUDA acceleration be used (default: False)
+        """
         return self._fpsubsim2.keep_dissimilar_molecules(smiles=smiles, fingerprint=fingerprint, threshold=threshold,
                                                          cuda=cuda)
 
     def keep_substructure_molecules(self, smiles: Union[str, List[str]]) -> PapyrusDataset:
+        """Keep samples whose molecular structures are substructures of any of the provided SMILES.
+
+        :param smiles: SMILES the molecular structures must not be substructures of
+        """
         return self._fpsubsim2.keep_substructure_molecules(smiles=smiles)
 
     def keep_not_substructure_molecules(self, smiles: Union[str, List[str]]) -> PapyrusDataset:
+        """Keep samples whose molecular structures are not substructures of any of the provided SMILES.
+
+        :param smiles: SMILES the molecular structures must not be substructures of
+        """
         return self._fpsubsim2.keep_not_substructure_molecules(smiles=smiles)
 
     def aggregate(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusDataset to a pandas DataFrame.
+
+        :param progress: should filtering progress be shown
+        :return: a pandas DataFrame of the filtered data.
+        """
         total = (-(-self.papyrus_params['num_rows'] // self.papyrus_params['chunksize'])
                  if self.papyrus_params['chunksize'] is not None
                  else None)
@@ -145,15 +241,35 @@ class PapyrusDataset:
                                          progress=progress, total=total)
 
     def agg(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusDataset to a pandas DataFrame (synonym of PapyrusDataset.aggregate).
+
+        :param progress: should filtering progress be shown
+        :return: a pandas DataFrame of the filtered data.
+        """
         return self.aggregate(progress=progress)
 
     def consume_chunks(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusDataset to a pandas DataFrame (synonym of PapyrusDataset.aggregate).
+
+        :param progress: should filtering progress be shown
+        :return: a pandas DataFrame of the filtered data.
+        """
         return self.aggregate(progress=progress)
 
     def to_dataframe(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusDataset to a pandas DataFrame (synonym of PapyrusDataset.aggregate).
+
+        :param progress: should filtering progress be shown
+        :return: a pandas DataFrame of the filtered data.
+        """
         return self.aggregate(progress=progress)
 
     def molecules(self, chunksize: Optional[int] = 1_000_000, progress: bool = False) -> PapyrusMoleculeSet:
+        """Get the structures of the molecules corresponding to the samples in the current PapyrusDataset.
+
+        :param chunksize: number of molecules to be loaded at once. To read without chunks (not recommended) set to None (default: 1_000_000).
+        :param progress: should progress of molecule aggregation be shown.
+        """
         ids = self.aggregate(progress=progress)['connectivity' if not self.papyrus_params['is3d'] else 'InChIKey'].unique()
         molecules = reader.read_molecular_structures(is3d=self.papyrus_params['is3d'],
                                                      version=self.papyrus_params['version'],
@@ -163,12 +279,24 @@ class PapyrusDataset:
         return PapyrusMoleculeSet(molecules, {**self.papyrus_params, 'chunksize': chunksize})
 
     def proteins(self, progress: bool = False) -> PapyrusProteinSet:
+        """Get the protein targets corresponding to the samples in the current PapyrusDataset.
+
+        :param progress: should progress of molecule aggregation be shown.
+        """
         ids = self.aggregate(progress=progress)['target_id'].unique()
         proteins = self.papyrus_protein_data[self.papyrus_protein_data.target_id.isin(ids)]
         return PapyrusProteinSet(proteins, self.papyrus_params,
                                  len(proteins))
 
     def match_rcsb_pdb(self, update: bool = True, progress: bool = False) -> PapyrusPDBProteinSet:
+        """Get the protein 3D structures from the RCSB Protein Data Bank of both protein targets and molecules
+        corresponding to the samples in the current PapyrusDataset .
+
+
+        :param update: should the local cache of PDB identifiers be updated (default: False).
+        :param progress: should progress of molecule aggregation be shown
+        :return: should progress of 3D structure aggregation be shown.
+        """
         total = (-(-self.papyrus_params['num_rows'] // self.papyrus_params['chunksize'])
                  if self.papyrus_params['chunksize'] is not None
                  else None)
@@ -206,6 +334,23 @@ class PapyrusDataset:
                force: bool = False,
                progress: bool = True,
                source_path: Optional[str] = None) -> None:
+        """Remove the data of the Papyrus dataset with multiple levels of deletion.
+
+        :param version: version to delete
+        :param remove_papyruspp: should Papyrus++ be removed
+        :param remove_bioactivities: should bioactivity data be removed
+        :param remove_proteins: should protein data be removed
+        :param remove_nostereo: should the stereochemistry-agnostic data be removed
+        :param remove_stereo: should the stereochemistry-aware data be removed
+        :param remove_structures: should molecular structures be removed
+        :param remove_descriptors: should molecular descriptors be removed
+        :param remove_other_files: should any additional file be removed
+        :param remove_version_root: should all the data of that version be removed
+        :param remove_papyrus_root: should all data all versions combined be removed
+        :param force: avoid asking for confirmation
+        :param progress: should deletion progress
+        :param source_path: folder containing the bioactivity dataset (default: pystow's home folder)
+        """
         download.remove_papyrus(outdir=source_path, version=version, papyruspp=remove_papyruspp,
                                 bioactivities=remove_bioactivities, proteins=remove_proteins,
                                 nostereo=remove_nostereo, stereo=remove_stereo, structures=remove_structures,
@@ -213,6 +358,32 @@ class PapyrusDataset:
                                 version_root=remove_version_root, papyrus_root=remove_papyrus_root,
                                 force=force, progress=progress)
 
+    def molecular_descriptors(self, desc_type: str, progress: bool = False) -> pd.DataFrame | Iterator[pd.DataFrame]:
+        """Obtain the molecular descriptors of the molecules in the current PapyrusMoleculeSet.
+
+        :param desc_type: type of descriptor to be obtained. One of {'mold2', 'mordred', 'cddd', 'fingerprint', 'moe', 'all'}
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecular descriptors.
+        """
+        ids = self.aggregate(progress)['connectivity' if not self.papyrus_params['is3d'] else 'InChIKey'].unique()
+        # Handle descriptors not yet downloaded
+        try:
+            return reader.read_molecular_descriptors(desc_type=desc_type,
+                                                     is3d=self.papyrus_params['is3d'],
+                                                     version=self.papyrus_params['version'],
+                                                     chunksize=self.papyrus_params['chunksize'],
+                                                     source_path=self.papyrus_params['source_path'],
+                                                     ids=ids,
+                                                     verbose=progress)
+        except FileNotFoundError:
+            download.download_papyrus(outdir=self.papyrus_params['source_path'],
+                                      version=self.papyrus_params['version'],
+                                      nostereo=not self.papyrus_params['is3d'], stereo=self.papyrus_params['is3d'],
+                                      only_pp=self.papyrus_params['plusplus'], structures=False,
+                                      descriptors=desc_type, progress=self.papyrus_params['download_progress'],
+                                      disk_margin=0.0)
+            return self.molecular_descriptors(desc_type, progress)
+
 class PapyrusDataFilter:
     """Collection of filters to be applied on a PapyrusDataset instance."""
 
@@ -426,6 +597,11 @@ class PapyrusMoleculeSet:
                                                 root_folder=self.papyrus_params['source_path'])
 
     def to_dataframe(self, progress: bool = False):
+        """Aggregate the data in a PapyrusMoleculeSet to a pandas DataFrame (synonym of `PapyrusMoleculeSet.aggregate`).
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecules.
+        """
         if isinstance(self.data, Iterator):
             return self.aggregate(progress=progress)
         return self.data
@@ -435,8 +611,14 @@ class PapyrusMoleculeSet:
             return f'{type(self).__name__}<iterator of molecules>'
         return f'{type(self).__name__}<{len(self.data)} molecules>'
 
-    def molecular_descriptors(self, desc_type: str, progress: bool = False) -> pd.DataFrame:
-        ids = self.aggregate(progress)['connectivity' if self.papyrus_params['is3d'] else 'InChIKey'].unique()
+    def molecular_descriptors(self, desc_type: str, progress: bool = False) -> pd.DataFrame | Iterator[pd.DataFrame]:
+        """Obtain the molecular descriptors of the molecules in the current PapyrusMoleculeSet.
+
+        :param desc_type: type of descriptor to be obtained. One of {'mold2', 'mordred', 'cddd', 'fingerprint', 'moe', 'all'}
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecular descriptors.
+        """
+        ids = self.aggregate(progress)['connectivity' if not self.papyrus_params['is3d'] else 'InChIKey'].unique()
         # Handle descriptors not yet downloaded
         try:
             return reader.read_molecular_descriptors(desc_type=desc_type,
@@ -456,6 +638,11 @@ class PapyrusMoleculeSet:
             return self.molecular_descriptors(desc_type, progress)
 
     def aggregate(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusMoleculeSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecules.
+        """
         total = (-(-self.num_rows // self.papyrus_params['chunksize'])
                  if self.papyrus_params['chunksize'] is not None
                  else None)
@@ -465,12 +652,56 @@ class PapyrusMoleculeSet:
                                          progress=progress, total=total)
 
     def agg(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusMoleculeSet to a pandas DataFrame (synonym of `PapyrusMoleculeSet.aggregate`).
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecules.
+        """
         return self.aggregate(progress=progress)
 
     def consume_chunks(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusMoleculeSet to a pandas DataFrame (synonym of `PapyrusMoleculeSet.aggregate`).
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the molecules.
+        """
         return self.aggregate(progress=progress)
 
-class PapyrusProteinSet:
+
+class ProteinSet(ABC):
+    """Abstract class."""
+
+    def protein_descriptors(self,
+                            desc_type: Union[str, prodec.Descriptor, prodec.Transform],
+                            progress: bool = False
+                            ) -> pd.DataFrame:
+        """Obtain the protein descriptors of the protein targets in the current PapyrusPDBProteinSet.
+
+        :param desc_type: type of protein descriptor to be obtained. Either 'unirep' or a `ProDEC.Descriptor` or `ProDEC.Transform`.
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the protein descriptors.
+        """
+        self.data = self.aggregate(progress)
+        ids = self.data['target_id'].unique()
+        try:
+            return reader.read_protein_descriptors(desc_type=desc_type,
+                                                   is3d=self.papyrus_params['is3d'],
+                                                   version=self.papyrus_params['version'],
+                                                   chunksize=self.papyrus_params['chunksize'],
+                                                   source_path=self.papyrus_params['source_path'],
+                                                   ids=ids,
+                                                   verbose=progress)
+        except FileNotFoundError:
+            download.download_papyrus(outdir=self.papyrus_params['source_path'],
+                                      version=self.papyrus_params['version'],
+                                      nostereo=not self.papyrus_params['is3d'], stereo=self.papyrus_params['is3d'],
+                                      only_pp=self.papyrus_params['plusplus'], structures=False,
+                                      descriptors=desc_type, progress=self.papyrus_params['download_progress'],
+                                      disk_margin=0.0)
+            return self.protein_descriptors(desc_type, progress)
+
+
+class PapyrusProteinSet(ProteinSet):
     def __init__(self, df: Union[pd.DataFrame, Iterator], papyrus_params: Dict, num_proteins: int):
         self.data = df
         self.papyrus_params = papyrus_params
@@ -482,6 +713,11 @@ class PapyrusProteinSet:
         return f'{type(self).__name__}<{len(self.data)} proteins>'
 
     def to_dataframe(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusProteinSet to a pandas DataFrame (synonym of `PapyrusProteinSet.aggregate`).
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the proteins.
+        """
         if isinstance(self.data, Iterator):
             total = (-(-self.num_rows // self.papyrus_params['chunksize'])
                      if self.papyrus_params['chunksize'] is not None
@@ -490,15 +726,31 @@ class PapyrusProteinSet:
         return self.data
 
     def aggregate(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the proteins.
+        """
         return self.to_dataframe(progress)
 
     def agg(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the proteins.
+        """
         return self.to_dataframe(progress=progress)
 
     def consume_chunks(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the proteins.
+        """
         return self.to_dataframe(progress=progress)
 
-class PapyrusPDBProteinSet:
+
+class PapyrusPDBProteinSet(ProteinSet):
 
     def __init__(self, df: Union[pd.DataFrame, Iterator], papyrus_params: Dict, num_proteins: int):
         self.data = df
@@ -506,6 +758,11 @@ class PapyrusPDBProteinSet:
         self.num_rows = num_proteins
 
     def to_dataframe(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusPDBProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the RCSB Protein Data Bank protein 3D structures.
+        """
         if isinstance(self.data, Iterator):
             total = (-(-self.num_rows // self.papyrus_params['chunksize'])
                      if self.papyrus_params['chunksize'] is not None
@@ -518,33 +775,27 @@ class PapyrusPDBProteinSet:
             return f'{type(self).__name__}<iterator of proteins structures>'
         return f'{type(self).__name__}<{len(self.data)} proteins structures>'
 
-    def protein_descriptors(self,
-                            desc_type: Union[str, prodec.Descriptor, prodec.Transform],
-                            progress: bool = False
-                            ) -> pd.DataFrame:
-        ids = self.aggregate(progress)['target_id'].unique()
-        try:
-            return reader.read_protein_descriptors(desc_type=desc_type,
-                                                   is3d=self.papyrus_params['is3d'],
-                                                   version=self.papyrus_params['version'],
-                                                   chunksize=self.papyrus_params['chunksize'],
-                                                   source_path=self.papyrus_params['source_path'],
-                                                   ids=ids,
-                                                   verbose=progress)
-        except FileNotFoundError:
-            download.download_papyrus(outdir=self.papyrus_params['source_path'],
-                                      version=self.papyrus_params['version'],
-                                      nostereo=not self.papyrus_params['is3d'], stereo=self.papyrus_params['is3d'],
-                                      only_pp=self.papyrus_params['plusplus'], structures=False,
-                                      descriptors=desc_type, progress=self.papyrus_params['download_progress'],
-                                      disk_margin=0.0)
-            return self.protein_descriptors(desc_type, progress)
 
     def aggregate(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusPDBProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the RCSB Protein Data Bank protein 3D structures.
+        """
         return self.to_dataframe(progress)
 
     def agg(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusPDBProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the RCSB Protein Data Bank protein 3D structures.
+        """
         return self.to_dataframe(progress=progress)
 
     def consume_chunks(self, progress: bool = False) -> pd.DataFrame:
+        """Aggregate the data in a PapyrusPDBProteinSet to a pandas DataFrame.
+
+        :param progress: should aggregation progress be shown
+        :return: a pandas DataFrame of the RCSB Protein Data Bank protein 3D structures.
+        """
         return self.to_dataframe(progress=progress)

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/src/papyrus_scripts/preprocess.py

@@ -94,11 +94,11 @@ def process_group(group, additional_columns: Optional[List[str]] = None):
     """Aggregate data from one group accordingly"""
     if (group.values[0] == group.values).all():  # If all values are equal, return first record
         group['pchembl_value_Mean'] = group['pchembl_value']
-        group['pchembl_value_StdDev'] = np.NaN
-        group['pchembl_value_SEM'] = np.NaN
+        group['pchembl_value_StdDev'] = np.nan
+        group['pchembl_value_SEM'] = np.nan
         group['pchembl_value_N'] = 1
         group['pchembl_value_Median'] = group['pchembl_value']
-        group['pchembl_value_MAD'] = np.NaN
+        group['pchembl_value_MAD'] = np.nan
         return group.iloc[:1, :]
     # Lambda: Return one value if all are the same
     listvals = lambda x: ';'.join(set(str(y) for y in x)) if (x.values[0] == x.values).all() else ';'.join(
@@ -174,8 +174,8 @@ def keep_source(data: Union[pd.DataFrame, PandasTextFileReader, Iterator], sourc
         # Allow processing of Papyrus++
         papyruspp = 'Activity_class' not in data.columns
         if papyruspp:
-            data['Activity_class'] = np.NaN
-            data['type_other'] = np.NaN
+            data['Activity_class'] = np.nan
+            data['type_other'] = np.nan
         # Keep trace of order of columns
         ordered_columns = data.columns.tolist()
         # Keep binary data associated to source
@@ -307,8 +307,8 @@ def keep_type(data: Union[pd.DataFrame, PandasTextFileReader, Iterator], activit
         # Allow processing of Papyrus++
         papyruspp = 'Activity_class' not in data.columns
         if papyruspp:
-            data['Activity_class'] = np.NaN
-            data['type_other'] = np.NaN
+            data['Activity_class'] = np.nan
+            data['type_other'] = np.nan
         # Transform activity_types to column names
         activity_types = [f"type_{types[i]}" for i in range(len(types)) if types_[i] in activity_types]
         # Columns with optional multiple values

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2/src/papyrus_scripts.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: papyrus_scripts
-Version: 2.1.0
+Version: 2.1.2
 Summary: A collection of scripts to handle the Papyrus bioactivity dataset
 Home-page: https://github.com/OlivierBeq/Papyrus-scripts
 Author: Olivier J. M. Béquignon - Brandon J. Bongers - Willem Jespers
@@ -17,7 +17,7 @@ Classifier: Programming Language :: Python :: 3.7
 Classifier: Programming Language :: Python :: 3.6
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy
+Requires-Dist: numpy>=2.0.0
 Requires-Dist: pandas
 Requires-Dist: rdkit
 Requires-Dist: requests
@@ -36,6 +36,7 @@ Requires-Dist: sphinx-autodoc-typehints; extra == "docs"
 Provides-Extra: testing
 Requires-Dist: pytest; extra == "testing"
 Requires-Dist: parameterized; extra == "testing"
+Dynamic: license-file
 
 # Papyrus-scripts
 
@@ -168,10 +169,10 @@ Different online servers host the Papyrus data based on release and ChEMBL versi
  
 | Papyrus version | ChEMBL version |                          Zenodo                           |                            4TU                            |
 |:---------------:|:--------------:|:---------------------------------------------------------:|:---------------------------------------------------------:|
-|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10944245) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
-|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/10943207) |                            :x:                            |
-|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7821775)  |                            :x:                            |
-|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787634) |                            :x:                            |
+|      05.4       |       29       | [:heavy_check_mark:](https://zenodo.org/records/10943992) | [:heavy_check_mark:](https://doi.org/10.4121/16896406.v2) | 
+|      05.5       |       30       | [:heavy_check_mark:](https://zenodo.org/records/7019873)  |                            :x:                            |
+|      05.6       |       31       | [:heavy_check_mark:](https://zenodo.org/records/7373213)  |                            :x:                            |
+|      05.7       |       34       | [:heavy_check_mark:](https://zenodo.org/records/13787633) |                            :x:                            |
 
 Precomputed molecular and protein descriptors along with molecular structures (2D for default set and 3D for low quality set with stereochemistry) are not available for version 05.4 from 4TU but are from Google Drive.
 

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/src/papyrus_scripts.egg-info/requires.txt

@@ -1,4 +1,4 @@
-numpy
+numpy>=2.0.0
 pandas
 rdkit
 requests

{papyrus_scripts-2.1.0 → papyrus_scripts-2.1.2}/tests/test_oop.py

@@ -181,8 +181,8 @@ class TestPapyrusDataset(unittest.TestCase):
         self.assertTrue(oop_data_agg.source.unique().item().lower().startswith('chembl'))
         self.assertTrue(oop_data_agg.type_IC50.dropna().astype(int).unique().item() == 0)
         self.assertTrue(oop_data_agg.type_EC50.dropna().astype(int).unique().item() == 0)
-        self.assertTrue(oop_data_agg.type_other.replace({'NA': np.NaN, 'NaN': np.NaN, 'nan': np.NaN})
-                        .dropna().empty or (oop_data_agg.type_other.replace({'NA': np.NaN, 'NaN': np.NaN, 'nan': np.NaN})
+        self.assertTrue(oop_data_agg.type_other.replace({'NA': np.nan, 'NaN': np.nan, 'nan': np.nan})
+                        .dropna().empty or (oop_data_agg.type_other.replace({'NA': np.nan, 'NaN': np.nan, 'nan': np.nan})
                         .dropna().astype(int).unique().item() == 0))
         self.assertEqual((oop_data_agg[['type_KD', 'type_Ki']]
                           .astype(int).